The present invention relates to a method of preparation of chemical compounds and, in particular, to a method of preparing combinatorial libraries of chemical compounds. The method is especially suitable for the preparation of natural and synthetic chemical compounds which are to be tested for activity as therapeutic agents, though it need not be used exclusively for this purpose. In addition to being used for the preparation of combinatorial libraries, the method of the present invention also facilitates easy identification of individual compounds, so that any compounds which show encouraging biological activity can be prepared on a larger scale for further analysis. By modifying the method of the present invention, it is possible to prepare individual compounds in pure form in a non-combinatorial format.
The synthesis and screening of combinatorial libraries is becoming increasingly important in the pharmaceutical industry as a means of drug "discovery". The major advantages of combinatorial chemistry are that it is faster and cheaper than orthodox methods. This makes it a much more effective technique in the quest to uncover new therapeutic agents, particularly in circumstances where there is little or no information available concerning the types of structures likely to show the desired activity.
The wider availability of solid-phase synthetic methods has also led to increased interest in combinatorial chemistry. Clearly, solution chemistry is unsuitable for a technique which aims to produce a multiplicity of new products together, since this does not allow physical separation between the different materials produced. The products are therefore likely be contaminated with excess reagents, by-products etc, leading to difficulties in separation and purification.
The preparation of combinatorial compound libraries typically involves a number of successive stages, each of which involves a chemical or enzymatic modification of an existing molecule. Most typically, this process involves the addition of a monomeric unit or other synthon to a growing sequence, or the modification of chemical functionality on the sequence. Conveniently, the sequence or growing chain of interest is attached to a solid support. By carrying out the desired series of synthetic steps on the bound starting material, and by altering the nature of the monomeric or other synthon units, the type of chemistry and the sequence of reactions, it is possible to prepare an enormous number of individual compounds in short time.
As indicated above, combinatorial methods entail a series of chemical steps with multiple choices of chemical reagents for each step. The complexity of the combinatorial library thus produced is determined by the product of the number of reagent choices for each step of the synthesis, which can be quite large. The problem which then arises is identification and characterisation of members of the library which display particular desired properties.
Various solutions have been proposed to deal with this: For example, members of the library can be synthesised in spatially segregated arrays. However, because of the extra burden which maintenance of segregation imposes, this approach tends to lead to relatively small libraries. Alternatively, in the so-called "multivalent synthesis" method, a library of moderate complexity is produced by pooling multiple choices of reagents during synthesis. If a pool is shown to have properties of interest, it is re-synthesised with progressively lower complexity until a single compound or class of compounds is identified having the desired property. The ultimate size of a library produced by this technique is inevitably restricted because of concentration effects which determine the limits of detection at which activity can be discerned.
The so-called "mix and split synthesis" method relies on combinatorial synthesis carried out on discrete solid particles such as minute resin beads. Through a protocol of mixing and separating beads at the end of each step in the synthetic sequence, populations of beads are generated to which are bound the products of specific reaction sequences. Inevitably, individual beads obtained from the final reaction step have different products attached, so that identification and characterisation of active materials is still a problem.
Fortunately, biologically active compounds show remarkable potency and receptor sites are highly selective, so it is possible to detect low concentrations of active compound amid an extensive background of inactive material using standard in vitro screening techniques.
Another drawback of the mix and split synthesis method is that some measure of over-representation and omission of individual compounds is inevitable because of the randomness introduced by the mixing and splitting steps.
To counteract the above problems of identification and characterisation, some workers have proposed co-synthesis of a sequencable tag which encodes the series of steps and reagents used during synthesis of respective constituents of the library. More recently, it has been proposed to use tagging molecules to encode both the step number and the chemical reagent used in a given step, as a binary record of the synthetic steps experienced by each bead. This technique undoubtedly adds to the complexity of operations carried out during development of a combinatorial library.
From the foregoing, it is apparent that known methods of preparing combinatorial libraries of chemical compounds suffer from two major drawbacks: Either the materials are prepared by maintaining segregation, with the inevitable consequence that only relatively small libraries are practicable, or the materials are prepared without segregation but in such minute quantities that characterisation is rendered very difficult.